Saturday, February 27, 2010

Earthquakes 'n sunshine

It's sunny in the mountains, and here I am stuck writing a twelve-page paper this weekend.

Mt. Saint Helens photo courtesy of USDA Forest Service, public domain

Actually I'm mostly tempted to rent a car and and head west to catch the small tsunami creeping up the coast. There are a number of harbors that will probably amplify the effect with good views at a safe distance.

Of course the wave from the Chilean earthquake is also a reminder that someday it will be our turn here in the Pacific Northwest, with 9.0 subduction quake unzipping the entire Cascadia zone from Northern California up into British Columbia.

On one hand the chances of this happening in our lifetime should be remote. Typically Cascadia tends to operate in clusters of four quakes spaced out three hundred years apart, followed by five hundred to a thousand years of silence.

However, this is nature and nature is uneven. While the 9.0 January 26th 1700 Cascadia quake was the fourth in the most recent cluster, groupings of five are not unknown. And there has been a fair amount of aseismic slippage taking place up around Vancouver Island in recent years. Additionally there is some recent evidence that the southern half of the fault ruptures more often than the northern half--normally every 280 years.

In Portland a full 9.0 quake along the length of the Cascadia zone would result in four to five minutes of long period shaking at 7.0 on the movement moment scale. Where normal quakes are violent and short lived, this one would involve drawn-out swaying over a prolonged period of time with a much greater amount of total energy being imputed into man-made structures.

That will be enough to take down older bridges and buildings, and isolate the region with landslides on all of the major highways and mountain passes.

Here in the city, high-rise buildings are a matter of particular concern. Normally these are the safest structures to be in during a quake. Tall steel-framed buildings sway gracefully and spread out the moment of impact in a wave across the length of the building rather than bearing the burden of each shake all at once. Four to five minutes, however, might be enough to put such buildings in resonance with the quake, meaning that the entire structure will be moving in synch with the ground.

Big Pink, pictured above, will at the very least end up shedding all of its glass and its marble facade during such a quake, and bury the streets around it four-feet deep in jagged debris. In the worst case scenario it will end up in forced resonance and vibrate itself to pieces.

Worst hit will be the region's un-reinforced masonry buildings. Bricks tend to work themselves loose and whole buildings come down.

Afterward, it will likely take a week or more for major relief to reach Portland. Not only would many of the region's bridges and mountain passes suffer damage, but the airport is built on an old flood plain, which will almost certainly liquefy underneath the runways. The Columbia River would also likely be un-navigable for a week or longer as its shipping channels would need to be re-dredged.

As bad as it might be here, Seattle and Vancouver, BC, will have it far worse.

While the oceanic plates that North America is overriding dive down steeply into the Earth's mantel along most of the coast, under the Puget Sound region the angle of attack is shallow. The means there is a broad band of locked contact between the North American Plate and the Juan De Fuca underneath it, and this zone extends inland. This overlap already leaves Seattle and Vancouver vulnerable to earthquakes from both the North American plate and--as in 2001--from the Juan de Fuca plate beneath it. In the case of a subduction zone quake--in which shaking will originate offshore for Northern California and Oregon--slippage and energy release will occur beneath the Olympic Mountains and Vancouver Island and subject Seattle and Vancouver to thirty-two times as much energy as Portland (8.0+).

The two northern cities are also largely built on jelly-like sedimentary runoff from the nearby mountains, which is likely to liquefy when shook. Additionally, the nearby Cascade Mountains will reflect a good deal of the quake's eastward moving energy back into the cities.

Recent research also shows an apparent link between Cascadia quakes and earthquakes on the San Andreas fault that runs across half the length of California. In other words it looks like Cascadia quakes often trigger San Andreas quakes to the south.

In the best case scenario, separate sections of the Cascadia fault will rupture in a series of 8.0+ quakes over the course of a decade or two. This will largely spare the inland cities, but repeatedly hammer the coast with large earthquakes and tsunamis.

All of this could take place today, or it could go down in another five hundred years from now. Part of the problem with estimating risk here is that we do not fully understand what the aseismic movements underneath Vancouver Island mean. Then there is the unexplained 1999 incident in which a large area of the zone quietly moved backwards several inches. The current best-guess estimate based on offshore sediment samples that contain a ten-thousand year record of quake-caused undersea landslides puts the fifty year chance of a full Cascadia quake at fourteen percent. However, based on the same evidence, some geologist believe that the southern half of the fault ruptures more often than the northern half, with an average of two hundred eighty years. If true, that means that Southern Oregon and Northern California are overdue for a Chilean-sized quake that would also transfer a large amount of stress into the northern San Andreas. The given odds for such an event in the next fifty years stands at eighty-percent.

Something to think about.

No comments: